JEEP HACKING CAN-C / CAN-IHS / UDS ! (Reverse Engineering)

[jmccorm]

In the meantime, can you ship me a copy of your config.txt so that I can rule out a bad configuration?
You are running on RPi4 hardware... right?

Thanks to some good weather, I stayed up tonight to work through the bulk of the problem. I believe it had to do with how I had previously configured my Raspberry Pi 3B. For a while, I could only get one path working at a time, and I believe I was tripping over what was stored via the persistent device numbering feature that can be enabled in raspi-config (reversing the natural order of the can device numbers).

I had everything wired the same, yet the CAN-IHS and CAN-C lines were going into the wrong places, so nothing worked at all (and it even made things lock up during boot). The problem couldn't be straightened out in /boot/config.txt, so in the end, it was a matter of physically swapping the cables.

Now I've got my latest instance of hardware and software out in the vehicle. It isn't yet connected to the RGB halos but the bits of color are actually twirling inside the computer. Still have to hook up the different voltages, but the voltage regulating software/hardware is working too. Things are looking up once again... ready to start moving forward. ?

Sponsored

 

[redracer]

Sorry, I had a busy evening. No, I am running a Pi 3B. Tho, I am still looking for the perfect Pi CM4 board to do a lot of what we want. Waveshare has one with can and wide input. I found another one as well with can and a screen that looks fun.
https://www.biqu.equipment/products...m_source=PR&utm_medium=3DPI&utm_campaign=Pad5

 

[jmccorm]

Sorry, I had a busy evening. No, I am running a Pi 3B. Tho, I am still looking for the perfect Pi CM4 board to do a lot of what we want. Waveshare has one with can and wide input. I found another one as well with can and a screen that looks fun.
https://www.biqu.equipment/products...m_source=PR&utm_medium=3DPI&utm_campaign=Pad5

Please, you do you... and keep pursuing that path. I'm interested in seeing something different here. More opportunities to consider! This is good.

 

[jmccorm]

QUESTION FOR ANYONE:
How much would a replacement glovebox set me back?

Things are starting to become too tight for nice neat work with the glovebox getting in the way.
I'm dealing with wires for:

• CAN-C --> Raspberry Pi
• CAN-IHS --> Raspberry Pi
• 12V BATT --> 5V Regulator --> Raspberry Pi (via Power2Go)
• 12V AUX3 --> Raspberry Pi (via Power2Go)
• 12V AUX4 --> 5V Regulator --> RGB LEDs AND Raspberry Pi (via Power2Go)
• Multiple signal and voltage grounds --> Everywhere

There's no real room for error back there. I think I'd like to shave away part of the glove box drawer and permanently mount some components on the inner surface of the glovebox. It seems like a decent enough idea, but how much am I talking? $100? $300?

 

[jeepster345]

QUESTION FOR ANYONE:
How much would a replacement glovebox set me back?

Things are starting to become too tight for nice neat work with the glovebox getting in the way.
I'm dealing with wires for:

• CAN-C --> Raspberry Pi
• CAN-IHS --> Raspberry Pi
• 12V BATT --> 5V Regulator --> Raspberry Pi (via Power2Go)
• 12V AUX3 --> Raspberry Pi (via Power2Go)
• 12V AUX4 --> 5V Regulator --> RGB LEDs AND Raspberry Pi (via Power2Go)
• Multiple signal and voltage grounds --> Everywhere

There's no real room for error back there. I think I'd like to shave away part of the glove box drawer and permanently mount some components on the inner surface of the glovebox. It seems like a decent enough idea, but how much am I talking? $100? $300?

Looks like a new one costs around $100.

 

[jmccorm]

Looks like a new one costs around $100.

NICE! Any tips you can provide on how I can hunt down answers to these kinds of question for myself? Certain tools at specific websites? Or all good connections and deep magic?

 

[jeepster345]

NICE! Any tips you can provide on how I can hunt down answers to these kinds of question for myself? Certain tools at specific websites? Or all good connections and deep magic?

You can just look up Jeep JL glovebox and find several options.

 

[redracer]

Ooo! I just received my esp32 CanLite board. This should be fun.

 

[redracer]

And I just figured out the command to reboot the radio.

cansend can0 7BF#0211010000000000

 

[jmccorm]

And I just figured out the command to reboot the radio.

cansend can0 7BF#0211010000000000

Congratulations! It looks like I've lost my UDS notes, so I'm a bit blind as to what you did (command 0x11, option 0x01). Was this just a UDS ECU Reset command that's sent to the message ID that's read by the radio? Simple as that? And the radio does a full and complete reset, or... ? Out of curiousity, if you tried an option 0x00, what did that do?

EDIT: You know, I bet there's a market for a low-cost device that has 8 built-in functions (such as this) which can be selected by a DIP switch. Set the DIP switches, hit the button, and it sends the appropriate command. Same turning off LED tail light diagnostics, other small things that should have been simple fixes.

 

[jmccorm]

I've finally finished off yet another foe that was keeping me offline. They had a regression in the zero2go installation script that plugged "dtoverlay=pi3-miniuart-bt" in the /boot/config.txt file. Normally, that might not be a problem, but I'm running on Raspberry Pi 4B hardware. The result was that the zero2go would instantly power my Raspberry Pi down every time it booted. That took some figuring out to solve. The more complex my solution stack gets, the more it's probably going to be like this. But at least I've got redundant environments now.

I've taken the doors off which has taken away some of the value of my in-vehicle Raspberry Pi (since it has all those remote-start options coded into it). I'll probably go back to my earlier effort in making dummy door plug connectors so that the Wrangler still thinks it's doors are attached (and will hopefully accept a remote start command once again).

Next step other than that is to get all my power feeds (AUX3, AUX4, BATTERY) squared away and fed into the zero2go (powering the Raspberry Pi) and then connecting the RGB output from the Raspberry Pi to the RGB Halos in the front of the vehicle. The wires have already been run between the glovebox and the engine compartment, it's a matter of power and wiring at this point.

I'm also trying to think of cool experiments or functions we could be developing right now. It would be awesome if one of us could loan the use of a high-end thinktool (or some other smart OBD-II / UDS diagnostic device) and figure out what functions we'd be able to copy from one of those. They'd be a goldmine!

Anyhow, I'm getting my mind back into the swing of things here. I've been set back for too long!

 

[WranglerAz]

NICE! Any tips you can provide on how I can hunt down answers to these kinds of question for myself? Certain tools at specific websites? Or all good connections and deep magic?

Send a PM on this site to Benny at AllMoparParts.com, he will track own the part numbers and cost. last order I did had a forum discount code.

 

[jmccorm]

Tonight I brought everything together for my RGB Halo Project and got it working. It was successful! And as I documented in my penultimate update, I made some updates which tied it back in with this project. The RGB Halos were CAN bus aware. In two ways!

A script monitors CAN-C message id 0x093 to read the current status of the gear selector. When the vehicle was in Park, the RGB halo lights operated in their native mode will full colors and features (such as flashing) unlocked. But the moment the vehicle goes into gear, it creates a flag in the filesystem which automatically tells the halo lights that they're being operated in street mode and they need to tone things down. That means white lights, variable intensities, but not colors and no flashing effects. The upshot is that the RGB halo lights can be wild in Park but legal on the streets and without requiring any intervention from the user to make it choose the right behavior at the right time. It just does it.

A second script monitors CAN-C message ID 0x291 to read the current status of the turn and hazard signals. Only while a turn (or hazard) signal was active, the matching RGB halo light would turn solid amber and remain that way as long as the signal remained lit. Once completed, the halo would switch back to it's normal behavior until needed once again. Enhanced turn signals.

Details and code are in the message I linked to. Two actual benefits brought to life from our investigation into the Wrangler's CAN bus network!

 

[jmccorm]

Next up is the AUX switch feature that I've been talking about for some time now:
An AUX switch that keeps running for a limited time after shutting down the vehicle!

Today I'm returning to the power situation on my Raspberry Pi 4B... and I'm hoping to pick up a feature that some people are sure to have wanted now and again over the years:
How about letting AUX4 automatically power on with ignition and then continue to run for 30 more minutes after the vehicle has been shut down?

Today's the day for the wiring which should make this possible with a bit of software.

As you know my Raspberry Pi 4B is currently powered through a Zero2Go Omini daughterboard which has three redundant power input channels. Currently we have dedicated 5V regulator on the AUX3 circuit (latched with ignition voltage). That feeds the Raspberry Pi and does nothing else. But that means that the Raspberry Pi is only operating while the engine is running. No turning engine, no smarts.

What I'll be adding today is a multipurpose 5V 5A regulator on AUX4. It'll be used to power the RGB Halos. It'll also be used to provide a diverse power feed to the Raspberry Pi (which under the right circumstances could allow it to continue to run and make decisions when AUX3 and ignition are off).

So today I'll be wiring up the 5V 25W regulator and feeding these two accessories with it.

After today's wiring, here's what I'm hoping we can make happen from the software side of things:

1. Vehicle starts normally, automatically powering AUX3 and booting the Raspberry Pi.
2. After booting, the Raspberry Pi will always reconfigure AUX4 (via CAN message) as latched battery voltage. This will power on the custom RGB halos. This also activates itself as a secondary power feed for the Raspberry Pi (which then allows it to continue running AFTER ignition and AUX3 have been powered down).
3. The Raspberry Pi constantly monitors ignition status (or AUX3 voltage levels), looking for when ignition has been shut down. In such a situation, AUX4 remains powered, and so does the Raspberry Pi.
4. The Raspberry Pi waits for 30 more minutes. It then Reconfigures AUX4 (via CAN message) as momentary battery voltage which powers down the AUX4 load as well as itself.

AUX3, AUX4, and the Raspberry Pi remains powered down until the vehicle is restarted.

The downside of this approach is that you're going to have to throw a lot of hardware and software at such as simple feature. For me, it'll be an incremental improvement. For a commercial solution, it should use a microcontroller.

So today is hardware and then after that I'm going to give this a whirl and see if I can have my RGB halos continue running for 30 minutes longer after I've parked the vehicle. (Perhaps shutting down sooner if it senses that I'm parked at home when it sees my local Wi-Fi router's advertisement.)

Here's hoping there's no new snags. My fear is that there may be an unknown hidden step to get power applied to the AUX4 circuit after you're configured it into a new state. If that ends up being the case, it could be a sizeable problem.

 

[redracer]

My fear is that there may be an unknown hidden step to get power applied to the AUX4 circuit after you're configured it into a new state. If that ends up being the case, it could be a sizeable problem.

This would break us into new territory... To get around this, we would have to get into the LIN bus. I have read up on this a bit, and it sounds like LIN bus are simpler than CAN bus, and it's truely just a simple serial data bus that we just need a receiver chip breakout to work with. This would be how we could directly interact with the aux switches, and also with the windows.

Sponsored